In this project, you'll generate your own Simpsons TV scripts using RNNs. You'll be using part of the Simpsons dataset of scripts from 27 seasons. The Neural Network you'll build will generate a new TV script for a scene at Moe's Tavern.
The data is already provided for you. You'll be using a subset of the original dataset. It consists of only the scenes in Moe's Tavern. This doesn't include other versions of the tavern, like "Moe's Cavern", "Flaming Moe's", "Uncle Moe's Family Feed-Bag", etc..
In [27]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper
# data_dir = '/Users/arasdar/arasdar-DL-git/data/simpsons/moes_tavern_lines.txt'
# data_dir = '/Users/arasdar/arasdar-DL-git/data/text_data/simpsons.txt'
data_dir = 'simpsons.txt'
text = helper.load_data(data_dir)
# Ignore notice, since we don't use it for analysing the data
text = text[81:]
In [28]:
view_sentence_range = (0, 10)
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np
print('Dataset Stats')
print('Roughly the number of unique words: {}'.format(len({word: None for word in text.split()})))
scenes = text.split('\n\n')
print('Number of scenes: {}'.format(len(scenes)))
sentence_count_scene = [scene.count('\n') for scene in scenes]
print('Average number of sentences in each scene: {}'.format(np.average(sentence_count_scene)))
sentences = [sentence for scene in scenes for sentence in scene.split('\n')]
print('Number of lines: {}'.format(len(sentences)))
word_count_sentence = [len(sentence.split()) for sentence in sentences]
print('Average number of words in each line: {}'.format(np.average(word_count_sentence)))
print()
print('The sentences {} to {}:'.format(*view_sentence_range))
print('\n'.join(text.split('\n')[view_sentence_range[0]:view_sentence_range[1]]))
The first thing to do to any dataset is preprocessing. Implement the following preprocessing functions below:
To create a word embedding, you first need to transform the words to ids. In this function, create two dictionaries:
vocab_to_intint_to_vocabReturn these dictionaries in the following tuple (vocab_to_int, int_to_vocab)
In [29]:
import numpy as np
import problem_unittests as tests
def create_lookup_tables(text):
"""
Create lookup tables for vocabulary
:param text: The text of tv scripts split into words
:return: A tuple of dicts (vocab_to_int, int_to_vocab)
"""
# TODO: Implement Function
# Split all the text into words - All the possible words collections/choices
#words = text.split() # This is not the unique words
#words = {word: None for word in text.split()}
#words = token_lookup()
from collections import Counter
# Count the freq of words in the text/collection of words
word_counts = Counter(text)
# Having counted the frequency of the words in collection, sort them from most to least/top to bottom/descendng
sorted_vocab = sorted(word_counts, key=word_counts.get, reverse=True)
# first enumerating for vocab to int
vocab_to_int = {words: ii for ii, words in enumerate(sorted_vocab)}
# into_to_vocab after enumerating through the sorted vocab
int_to_vocab = {ii: words for words, ii in vocab_to_int.items()}
# return the output results: a tuple of dicts(vocab_to_int, int_to_vocab)
# return dicts(vocab_to_int, int_to_vocab)
return vocab_to_int, int_to_vocab
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_create_lookup_tables(create_lookup_tables)
We'll be splitting the script into a word array using spaces as delimiters. However, punctuations like periods and exclamation marks make it hard for the neural network to distinguish between the word "bye" and "bye!".
Implement the function token_lookup to return a dict that will be used to tokenize symbols like "!" into "||Exclamation_Mark||". Create a dictionary for the following symbols where the symbol is the key and value is the token:
This dictionary will be used to token the symbols and add the delimiter (space) around it. This separates the symbols as it's own word, making it easier for the neural network to predict on the next word. Make sure you don't use a token that could be confused as a word. Instead of using the token "dash", try using something like "||dash||".
In [30]:
def token_lookup():
"""
Generate a dict to turn punctuation into a token.
:return: Tokenize dictionary where the key is the punctuation and the value is the token
"""
# TODO: Implement Function
# Replace punctuation with tokens so we can use them in our model
token_dict = {'.': '||Period||',
',': '||Comma||',
'"': '||Quotation_Mark||',
';': '||Semicolon||',
'!': '||Exclamation_Mark||',
'?': '||Question_Mark||',
'(': '||Left_Parentheses||',
')': '||Right_Parentheses||',
'--': '||Dash||',
'\n': '||Return||'}
#token_dict.items() # to show it
return token_dict
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_tokenize(token_lookup)
In [31]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
# Preprocess Training, Validation, and Testing Data
helper.preprocess_and_save_data(data_dir, token_lookup, create_lookup_tables)
In [32]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper
import numpy as np
import problem_unittests as tests
int_text, vocab_to_int, int_to_vocab, token_dict = helper.load_preprocess()
In [33]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf
# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer'
print('TensorFlow Version: {}'.format(tf.__version__))
# Check for a GPU
if not tf.test.gpu_device_name():
warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
Implement the get_inputs() function to create TF Placeholders for the Neural Network. It should create the following placeholders:
name parameter.Return the placeholders in the following the tuple (Input, Targets, LearingRate)
In [34]:
def get_inputs():
"""
Create TF Placeholders for input, targets, and learning rate.
:return: Tuple (input, targets, learning rate)
"""
# TODO: Implement Function
input = tf.placeholder(dtype=tf.int32, shape=[None, None], name='input')
targets = tf.placeholder(dtype=tf.int32, shape=[None, None], name='targets')
lr = tf.placeholder(dtype=tf.float32, shape=None, name='learning_rate')
return input, targets, lr
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_inputs(get_inputs)
It’s almost too easy to use the standard GRU or LSTM cells, so let’s define our own RNN cell. Here’s a random idea that may or may not work: starting with a GRU cell, instead of taking a single transformation of its input, we enable it to take a weighted average of multiple transformations of its input. The idea is that we might benefit from treat the input differently in different scenarios (e.g., we may want to treat verbs differently than nouns).
To write the custom cell, we need to extend tf.nn.rnn_cell.RNNCell. Specifically, we need to fill in 3 abstract methods and write an init method. First, let’s start with a GRU cell, adapted from Tensorflow’s implementation:
In [35]:
# class GRUCell(tf.nn.rnn_cell.RNNCell):
class GRUCell(tf.contrib.rnn.RNNCell):
"""Gated Recurrent Unit cell (cf. http://arxiv.org/abs/1406.1078)."""
def __init__(self, num_units):
self._num_units = num_units
@property
def state_size(self):
return self._num_units
@property
def output_size(self):
return self._num_units
def __call__(self, inputs, state, scope=None):
with tf.variable_scope(scope or type(self).__name__): # "GRUCell"
with tf.variable_scope("Gates"): # Reset gate and update gate.
# We start with bias of 1.0 to not reset and not update.
# ru = tf.nn.rnn_cell._linear([inputs, state],
# 2 * self._num_units, True, 1.0)
ru = tf.contrib.layers.linear([inputs, state],
2 * self._num_units, True, 1.0)
ru = tf.nn.sigmoid(ru)
r, u = tf.split(1, 2, ru)
with tf.variable_scope("Candidate"):
# c = tf.nn.tanh(tf.nn.rnn_cell._linear([inputs, r * state],
# self._num_units, True))
c = tf.nn.tanh(tf.contrib.layers.linear([inputs, r * state],
self._num_units, True))
new_h = u * state + (1 - u) * c
return new_h, new_h
We modify the init method to take a parameter at initialization, which will determine the number of transformation matrices. def init(self, num_units, num_weights): self._num_units = num_units self._num_weights = num_weights Then, we modify the Candidate variable scope of the call method to do a weighted average as shown below (note that all of the matrices are created as a single variable and then split into multiple tensors):
In [36]:
# class CustomCell(tf.nn.rnn_cell.RNNCell):
class CustomRNNCell(tf.contrib.rnn.RNNCell):
"""Gated Recurrent Unit cell (cf. http://arxiv.org/abs/1406.1078)."""
def __init__(self, num_units, num_weights):
self._num_units = num_units
self._num_weights = num_weights
@property
def state_size(self):
return self._num_units
@property
def output_size(self):
return self._num_units
def __call__(self, inputs, state, scope=None):
with tf.variable_scope(scope or type(self).__name__): # "GRUCell"
with tf.variable_scope("Gates"): # Reset gate and update gate.
# We start with bias of 1.0 to not reset and not update.
# ru = tf.nn.rnn_cell._linear([inputs, state],
# 2 * self._num_units, True, 1.0)
ru = tf.contrib.layers.linear([inputs, state],
2 * self._num_units, True, 1.0)
ru = tf.nn.sigmoid(ru)
r, u = tf.split(1, 2, ru)
with tf.variable_scope("Candidate"):
# lambdas = tf.nn.rnn_cell._linear([inputs, state], self._num_weights, True)
# tf.contrib.layers.linear
lambdas = tf.contrib.layers.linear([inputs, state], self._num_weights, True)
lambdas = tf.split(1, self._num_weights, tf.nn.softmax(lambdas))
Ws = tf.get_variable("Ws",
shape = [self._num_weights, inputs.get_shape()[1], self._num_units])
Ws = [tf.squeeze(i) for i in tf.split(0, self._num_weights, Ws)]
candidate_inputs = []
for idx, W in enumerate(Ws):
candidate_inputs.append(tf.matmul(inputs, W) * lambdas[idx])
Wx = tf.add_n(candidate_inputs)
b = tf.contrib.layers.linear([r * state], self._num_units, True, scope="second")
# c = tf.nn.tanh(Wx + tf.nn.rnn_cell._linear([r * state],
# self._num_units, True, scope="second"))
c = tf.nn.tanh(Wx+b)
new_h = u * state + (1 - u) * c
return new_h, new_h
Stack one or more BasicLSTMCells in a MultiRNNCell.
rnn_sizezero_state() functiontf.identity()Return the cell and initial state in the following tuple (Cell, InitialState)
In [54]:
def get_init_cell(batch_size, rnn_size):
"""
Create an RNN Cell and initialize it.
:param batch_size: Size of batches
:param rnn_size: Size of RNNs
:return: Tuple (cell, initialize state)
"""
# TODO: Implement Function
# The number of LSTM cells/ Memory cells in one layer for RNN
# rnn = tf.contrib.rnn.BasicLSTMCell(rnn_size, reuse=tf.get_variable_scope().reuse) # rnn_size==LSTM_size??
# rnn = GRUCell(num_units=rnn_size)
# # # Adding Dropout NOT needed/ Not Asked
# # keep_prob = 1.0 # Drop out probability
# # drop = tf.contrib.rnn.DropoutWrapper(rnn, keep_prob) #output_keep_prop=
# # Stacking up multiple LSTM layers for DL
# rnn_layers = 1 # layers
# cell = tf.contrib.rnn.MultiRNNCell([rnn] * rnn_layers)
# # Please create a new instance of the cell if you would like it to use a different set of weights.
# # If before you were using: MultiRNNCell([BasicLSTMCell(...)] * num_layers),
# # change to: MultiRNNCell([BasicLSTMCell(...) for _ in range(num_layers)]).
# # cells = [rnn for _ in range(rnn_layers)]
# # cell = tf.contrib.rnn.MultiRNNCell(cells=cells)
cell = tf.contrib.rnn.NASCell(num_units=rnn_size, reuse=tf.get_variable_scope().reuse)
lstm_cell = tf.contrib.rnn.DropoutWrapper(cell, output_keep_prob=0.8)
# num_layers=1
# rnn_cells = tf.contrib.rnn.MultiRNNCell([lstm_cell for _ in range(num_layers)], state_is_tuple = True)
# Initializing the cell state using zero_state()
initial_state = rnn_cells.zero_state(batch_size, tf.float32)
initial_state = tf.identity(input=initial_state, name='initial_state')
return lstm_cell, initial_state
# Aras: Already implemented in sentiment network
# lstm_size = 256
# lstm_layers = 1
# batch_size = 500
# learning_rate = 0.001
# with graph.as_default():
# # Your basic LSTM cell
# lstm = tf.contrib.rnn.BasicLSTMCell(lstm_size)
# # Add dropout to the cell
# drop = tf.contrib.rnn.DropoutWrapper(lstm, output_keep_prob=keep_prob)
# # Stack up multiple LSTM layers, for deep learning
# cell = tf.contrib.rnn.MultiRNNCell([drop] * lstm_layers)
# # Getting an initial state of all zerosoutput_keep_prop
# initial_state = cell.zero_state(batch_size, tf.float32)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_init_cell(get_init_cell)
You created a RNN Cell in the get_init_cell() function. Time to use the cell to create a RNN.
tf.nn.dynamic_rnn()tf.identity()Return the outputs and final_state state in the following tuple (Outputs, FinalState)
In [114]:
def build_rnn(inputs, batch_size, rnn_size): #cell,
"""
Create a RNN using a RNN Cell
:param cell: RNN Cell
:param inputs: Input text data
:return: Tuple (Outputs, Final State)
"""
# TODO: Implement Function
# cell = tf.contrib.rnn.NASCell(num_units=rnn_size, reuse=tf.get_variable_scope().reuse)
# cell = tf.contrib.rnn.DropoutWrapper(cell=cell, output_keep_prob=0.8)
# num_layers=1
# cell = tf.contrib.rnn.MultiRNNCell([cell for _ in range(num_layers)], state_is_tuple = True)
# The number of LSTM cells/ Memory cells in one layer for RNN
rnn = tf.contrib.rnn.BasicLSTMCell(rnn_size, reuse=tf.get_variable_scope().reuse) # rnn_size==LSTM_size??
# Stacking up multiple LSTM layers for DL
rnn_layers = 1 # layers
cell = tf.contrib.rnn.MultiRNNCell([rnn] * rnn_layers)
# Initializing the cell state using zero_state()
initial_state = cell.zero_state(batch_size, tf.float32)
initial_state = tf.identity(input=initial_state, name='initial_state')
outputs, final_state = tf.nn.dynamic_rnn(cell=cell, inputs=inputs, initial_state=None, dtype=tf.float32)
final_state = tf.identity(input=final_state, name='final_state')
# Returning the outputs and the final_state
return outputs, final_state, initial_state
# Aras: Implementation from Sentiment_RNN_Solution.ipynb
# with graph.as_default():
# outputs, final_state = tf.nn.dynamic_rnn(cell, embed,
# initial_state=initial_state)
# """
# DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
# """
# tests.test_build_rnn(build_rnn)
In [115]:
def get_embed(input_data, vocab_size, embed_dim):
"""initial_state
Create embedding for <input_data>.
:param input_data: TF placeholder for text input.
:param vocab_size: Number of words in vocabulary.
:param embed_dim: Number of embedding dimensions
:return: Embedded input.
"""
# TODO: Implement Function
# Size of embedding vectors (number of units in the emdding layer)
# embedding = tf.Variable(tf.random_uniform((vocab_size, embed_dim), -1, 1))
embedding = tf.Variable(tf.random_uniform(shape=[vocab_size, embed_dim], minval=-1, maxval=1, dtype=tf.float32,
seed=None, name=None))
# tf.random_normal(mean=1.0, size/shape=[], stddev=0.1)
# tf.random_normal(shape=[vocab_size/n_words, embed_size/embed_dim], mean=0.0, stddev=1.0,
#dtype=tf.float32, seed=None, name=None)
embed = tf.nn.embedding_lookup(embedding, input_data)
return embed
# # Embedding implementation from Sentiment_RNN_solution.ipynb
# # Size of the embedding vectors (number of units in the embedding layer)
# embed_size = 300
# with graph.as_default():
# embedding = tf.Variable(tf.random_uniform((n_words, embed_size), -1, 1))
# embed = tf.nn.embedding_lookup(embedding, inputs_)tf.nn.dynamic_rnn
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_embed(get_embed)
Apply the functions you implemented above to:
input_data using your get_embed(input_data, vocab_size, embed_dim) function.cell and your build_rnn(cell, inputs) function.vocab_size as the number of outputs.Return the logits and final state in the following tuple (Logits, FinalState)
In [116]:
def build_nn(rnn_size, input_data, vocab_size):
"""
Build part of the neural network
:param cell: RNN cell
:param rnn_size: Size of rnns
:param input_data: Input data
:param vocab_size: Vocabulary size
:return: Tuple (Logits, FinalState)
"""
# TODO: Implement Function
# embedding layer: def get_embed(input_data, vocab_size, embed_dim):
embed = get_embed(input_data=input_data, vocab_size=vocab_size, embed_dim=rnn_size)
# build rnn: def build_rnn(cell, inputs):
outputs, final_state, initial_state = build_rnn(inputs=embed, batch_size=vocab_size, rnn_size=rnn_size)
# Apply a fully connected layer with a linear activation and vocab_size as the number of outputs.
logits = tf.contrib.layers.fully_connected(inputs=outputs, num_outputs=vocab_size, activation_fn=None)
return logits, final_state, initial_state
# """
# DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
# """
# tests.test_build_nn(build_nn)
Implement get_batches to create batches of input and targets using int_text. The batches should be a Numpy array with the shape (number of batches, 2, batch size, sequence length). Each batch contains two elements:
[batch size, sequence length][batch size, sequence length]If you can't fill the last batch with enough data, drop the last batch.
For exmple, get_batches([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15], 2, 3) would return a Numpy array of the following:
[
# First Batch
[
# Batch of Input
[[ 1 2 3], [ 7 8 9]],
# Batch of targets
[[ 2 3 4], [ 8 9 10]]
],
# Second Batch
[
# Batch of Input
[[ 4 5 6], [10 11 12]],
# Batch of targets
[[ 5 6 7], [11 12 13]]
]
]
In [117]:
def get_batches(int_text, batch_size, seq_length):
"""
Return batches of input and target
:param int_text: Text with the words replaced by their ids
:param batch_size: The size of batch
:param seq_length: The length of sequence
:return: Batches as a Numpy array
"""
# TODO: Implement Function
# calculating the batch length, i.e. number words in a batch
batch_length = batch_size*seq_length # Remeber: batch_length != batch_size
# Number of bacthes in the give text with word IDs
num_batches = len(int_text)// batch_length
if (len(int_text)//batch_length) == (len(int_text)/batch_length):
num_batches -= 1
# preparing the numpy array first which is going to be returned/outputed
batches = np.zeros([num_batches, 2, batch_size, seq_length])
# number of words in the text (out dataset)
# get rid of the rest of the text which can fully be included in a batch based on the batch size
int_text = int_text[:(num_batches*batch_length)+1] # incremented one for the IO sequences/seq2seq learning
# Now based on the txt_size, batch_size, and seq_size/length, we should start getting the batches stochastically
#for batch_index/b_idx in range(start=0, stop=len(int_text), step=batch_size):
for batch_idx in range(0, num_batches, 1):
batch_slice = int_text[batch_idx*batch_length:(batch_idx+1)*batch_length+1]
# Slicing up the sequences inside a batch
#for seq_index/s_idx in range(start=0, stop=len(batch[??]), step=seq_length): # remember each sequence has two seq: input & output
for seq_idx in range(0, batch_size, 1):
batches[batch_idx, 0, seq_idx] = batch_slice[seq_idx*seq_length:(seq_idx+1)*seq_length]
batches[batch_idx, 1, seq_idx] = batch_slice[seq_idx*seq_length+1:((seq_idx+1)*seq_length)+1]
return batches
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_batches(get_batches)
Tune the following parameters:
num_epochs to the number of epochs.batch_size to the batch size.rnn_size to the size of the RNNs.seq_length to the length of sequence.learning_rate to the learning rate.show_every_n_batches to the number of batches the neural network should print progress.
In [118]:
# Number of Epochs
num_epochs = 100 # depends on how fast the system is and how long we can wait to see the results
# Batch Size
batch_size = 64 # depends on the memory, num seq per batch
# RNN Size
rnn_size = 128 # Pixel and int/8 Bit
# Sequence Length
seq_length = 64 # the same as RNN width size/ number of mem cells in one layer
# Learning Rate
learning_rate = 0.01
# Show stats for every n number of batches
show_every_n_batches = 16 # 2^4 show every 16 batches learning/training
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
save_dir = './save'
In [119]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from tensorflow.contrib import seq2seq
train_graph = tf.Graph()
with train_graph.as_default():
vocab_size = len(int_to_vocab)
input_text, targets, lr = get_inputs()
input_data_shape = tf.shape(input_text)
# cell, initial_state = get_init_cell(input_data_shape[0], rnn_size)
# logits, final_state = build_nn(cell, rnn_size, input_text, vocab_size)
logits, final_state, initial_state = build_nn(input_data=input_text, rnn_size=rnn_size, vocab_size=vocab_size)
# Probabilities for generating words
probs = tf.nn.softmax(logits, name='probs')
# Loss function
cost = seq2seq.sequence_loss(
logits,
targets,
tf.ones([input_data_shape[0], input_data_shape[1]]))
# Optimizer
optimizer = tf.train.AdamOptimizer(lr)
# Gradient Clipping
gradients = optimizer.compute_gradients(cost)
capped_gradients = [(tf.clip_by_value(grad, -1., 1.), var) for grad, var in gradients]
train_op = optimizer.apply_gradients(capped_gradients)
Train the neural network on the preprocessed data. If you have a hard time getting a good loss, check the forms to see if anyone is having the same problem.
In [120]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
batches = get_batches(int_text, batch_size, seq_length)
with tf.Session(graph=train_graph) as sess:
sess.run(tf.global_variables_initializer())
for epoch_i in range(num_epochs):
state = sess.run(initial_state, {input_text: batches[0][0]})
for batch_i, (x, y) in enumerate(batches):
feed = {
input_text: x,
targets: y,
initial_state: state,
lr: learning_rate}
train_loss, state, _ = sess.run([cost, final_state, train_op], feed)
# Show every <show_every_n_batches> batches
if (epoch_i * len(batches) + batch_i) % show_every_n_batches == 0:
print('Epoch {:>3} Batch {:>4}/{} train_loss = {:.3f}'.format(
epoch_i,
batch_i,
len(batches),
train_loss))
# Save Model
saver = tf.train.Saver()
saver.save(sess, save_dir)
print('Model Trained and Saved')
In [25]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
# Save parameters for checkpoint
helper.save_params((seq_length, save_dir))
In [26]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import tensorflow as tf
import numpy as np
import helper
import problem_unittests as tests
_, vocab_to_int, int_to_vocab, token_dict = helper.load_preprocess()
seq_length, load_dir = helper.load_params()
Get tensors from loaded_graph using the function get_tensor_by_name(). Get the tensors using the following names:
Return the tensors in the following tuple (InputTensor, InitialStateTensor, FinalStateTensor, ProbsTensor)
In [29]:
def get_tensors(loaded_graph):
"""
Get input, initial state, final state, and probabilities tensor from <loaded_graph>
:param loaded_graph: TensorFlow graph loaded from file
:return: Tuple (InputTensor, InitialStateTensor, FinalStateTensor, ProbsTensor)
"""
# TODO: Implement Function
input = loaded_graph.get_tensor_by_name(name='input:0')
initial_state = loaded_graph.get_tensor_by_name(name='initial_state:0')
final_state = loaded_graph.get_tensor_by_name(name='final_state:0')
probs = loaded_graph.get_tensor_by_name(name='probs:0')
return input, initial_state, final_state, probs
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_tensors(get_tensors)
In [30]:
def pick_word(probabilities, int_to_vocab):
"""
Pick the next word in the generated text
:param probabilities: Probabilites of the next word
:param int_to_vocab: Dictionary of word ids as the keys and words as the values
:return: String of the predicted word
"""
# TODO: Implement Function
# extractin the string/words out of the int_to_vocab.items
words = np.array([words for ids, words in int_to_vocab.items()])
# The generated random samples = numpy.random.choice(a, size=None, replace=True, p=None)¶
random_word = np.random.choice(a = words, size=None, replace=True, p=probabilities)
return random_word
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_pick_word(pick_word)
In [31]:
gen_length = 200
# homer_simpson, moe_szyslak, or Barney_Gumble
prime_word = 'moe_szyslak'
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
loaded_graph = tf.Graph()
with tf.Session(graph=loaded_graph) as sess:
# Load saved model
loader = tf.train.import_meta_graph(load_dir + '.meta')
loader.restore(sess, load_dir)
# Get Tensors from loaded model
input_text, initial_state, final_state, probs = get_tensors(loaded_graph)
# Sentences generation setup
gen_sentences = [prime_word + ':']
prev_state = sess.run(initial_state, {input_text: np.array([[1]])})
# Generate sentences
for n in range(gen_length):
# Dynamic Input
dyn_input = [[vocab_to_int[word] for word in gen_sentences[-seq_length:]]]
dyn_seq_length = len(dyn_input[0])
# Get Prediction
probabilities, prev_state = sess.run(
[probs, final_state],
{input_text: dyn_input, initial_state: prev_state})
pred_word = pick_word(probabilities[dyn_seq_length-1], int_to_vocab)
gen_sentences.append(pred_word)
# Remove tokens
tv_script = ' '.join(gen_sentences)
for key, token in token_dict.items():
ending = ' ' if key in ['\n', '(', '"'] else ''
tv_script = tv_script.replace(' ' + token.lower(), key)
tv_script = tv_script.replace('\n ', '\n')
tv_script = tv_script.replace('( ', '(')
print(tv_script)
It's ok if the TV script doesn't make any sense. We trained on less than a megabyte of text. In order to get good results, you'll have to use a smaller vocabulary or get more data. Luckly there's more data! As we mentioned in the begging of this project, this is a subset of another dataset. We didn't have you train on all the data, because that would take too long. However, you are free to train your neural network on all the data. After you complete the project, of course.
When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_tv_script_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.